Brain injury during cardiac surgery results primarily from cerebral embolism and/or reduced cerebral blood flow (CBF). The latter is of particular concern for the growing number of surgical patients who are aged and/or who have cerebral vascular disease. Normally, CBF is physiologically autoregulated (or kept constant) within a range of blood pressures allowing for stable cerebral O2 supply commensurate with metabolic demands. Cerebral autoregulation is impaired in patients undergoing cardiac surgery who have cerebral vascular disease and in many others due to other conditions. This could lead to brain injury since current practices of targeting low mean arterial blood pressure empirically (usually 50-70 mmHg) during cardiopulmonary bypass may expose patients with impaired cerebral autoregulation to cerebral hypoperfusion. The hypothesis of this proposal is that targeting mean arterial pressure during cardiopulmonary bypass to a level above an individual's lower autoregulatory threshold reduces the risk for brain injury in patients undergoing cardiac surgery. Monitoring of cerebral autoregulation will be performed in real time using software that continuously compares the relation between arterial blood pressure and CBF velocity of the middle cerebral artery measured with transcranial Doppler and with cerebral oximetry measured with near infrared spectroscopy. The primary end-point of the study will be a comprehensive composite outcome of clinical stroke, cognitive decline, and/or new ischemic brain lesions detected with diffusion weighted magnetic resonance (MR) imaging. Autoregulation is mediated by reactivity of cerebral resistance vessels. A secondary aim of this proposal is to evaluate whether near infrared reflectance spectroscopy can be used to trend changes in cerebral blood volume and provide a reliable monitor of vascular reactivity (the hemoglobin volume index). Assessments for extra-cranial and intra-cranial arterial stenosis will be performed using MR angiography to control for this potential confounding variable in the analysis. Finally, an additional aim of the study will be to assess whether preoperative transcranial Doppler examination of major cerebral arteries can identify patients who are prone to the composite neurological end-point. Near infrared oximetry is non-invasive, continuous, requires little care- giver intervention and, thus, could be widely used to individualize patient blood pressure management during surgery. Brain injury from cardiac surgery is an important source of operative mortality, prolonged hospitalization, increased health care expenditure, and impaired quality of life. Developing strategies to reduce the burden of this complication has wide public health implications and is within the mission of the NHLBI.